View metadata, citation and similar papers at core.ac.uk brought to you by CORE

provided by UC Research Repository

ANTIBIOTIC COMPOUNDS FROM NEW ZEALAND PLANTS: METHYL HAEMATOMMATE, AN ANTI-FUNGAL AGENT FROM RAMULOSUM

BJ. HICKEY, AJ. LUMSDEN, A.L.J. COLE, & J.R.L. WALKER1

Department of Plant and Microbial Sciences, University of Canterbury, Christchurch, New Zealand.

(Received 24 April, 1990; revised and accepted 27 May, 1990)

ABSTRACT

Hickey, BJ., Lumsden, AJ., Cole, A.LJ & Walker, J.R.L. (1990). Antibiotic compounds from New Zealand plants: Methyl haematommate, an anti-fungal agent from Stereocaulon ramulosum. New Zealand Natural Sciences 17: 49-53.

Methyl haematommate (methyl, 3-aldo-2:4-dihydroxy-6-methylbenzoate) was isolated from the Stere­ ocaulon ramulosum. This compound showed marked antibiotic activity against dermatophytic fungi: treated cells showed gross morphological changes and marked differences in their pattern of sterol metabolism.

KEYWORDS: Stereocaulon ramulosum - lichen - anti-mycotic - dermatophytes - methyl haematommate - steroid biosynthesis.

INTRODUCTION antibiotic activity: methanol was found to be the best extractant. Subsequently 25 g samples were Earlier studies in this laboratory (Calder et extracted with three successive 150 ml portions al. 1986) found that extracts from a number of of methanol for 12 h in a Soxhlet apparatus. New Zealand's unique plants exhibited different These were then concentrated in a rotary vac­ types of antibiotic activity. Here we wish to re­ uum evaporator care being taken to avoid total port the isolation and identification of an anti- drying of the extract (Fraction I). dermatophyte compound from the lichen Stere­ ocaulon ramulosum (SW.) Rausch. CHROMATOGRAPHIC SEPARATIONS Thin-layer chromatography (TLC) was car­ MATERIALS AND METHODS ried out on silica gel coated sheets developed with Culberson's solvent system C (toluene, ace­ Lichen was collected from various sites on tic acid, 85:45) (Culberson & Kristinson 1970). the Port Hills, near Christchurch, N.Z. and a Lichen constituents were visualised by examina­ voucher specimen was deposited in the Depart­ tion under UV light (254 and 365 nm), exposure ment's herbarium. Samples, cleaned of extrane­ to iodine vapour or spraying with 10% (v/v) sul­ ous moss and soil, were preserved by freeze- phuric acid followed by heating at 90°C to bring drying, and stored in sealed containers at 4°C. about colour development. Fungal steroids were extracted from freeze- EXTRACTION PROCEDURES dried mycelium with dimethyl sulphoxide-hexane Freeze-dried lichen was ground to a coarse or chloroform-methanol and then analysed by powder in a small hammer mill, and 1 g samples TLC on silver nitrate-impregnated silica gel extracted with a range of different solvents to plates (Parkes et al. 1985) developed with ben­ determine the optimum extractant for maximum zene-ethyl acetate (1:2 v/v) (Lisboa 1969). Cho­ lesterol, ergosterol and squalene were included To whom correspondence should be addressed as standards. Steroids were visualised by spray- 50 New Zealand Natural Sciences 17 (1990) ing the TLC plates with Liebermann-Burchard peaks were located at 221, 252, 292 and 382 nm. reagent (Lisboa 1969) or concentrated sulphuric These observations suggested the presence of a acid -methanol (1:1) followed by heating at 90°C phenolic compound. This was confirmed by for 10 min (Stahl 1969). NMR spectroscopy which suggested that Com­ pound I was a small, 10C, multiple-substituted, ANTIBIOTIC ASSAY phenolic compound. The NMR chemical shifts For the routine bioassay of antibiotic activity, indicated the presence of only one aromatic pro­ test samples were dried onto 6 mm antibiotic ton (SH 6.28), two phenolic groups (SH 12.41 assay (AA) discs and placed on agar plates previ­ and 12.88), a formyl proton (SH 10.33) with as­ ously seeded with a suspension of the test organ­ sociated carbonyl (SC 193.67) plus a further es­ ism. Dermatophyte fungi were all recent isolates ter linkage carbonyl (SC 171.83), a methoxyl from local clinical laboratories. group (SH 3.96, SC 52.33) and a methyl group Whole TLC plates (50 x 200 mm) were bio­ (SH 2.53). The 13C-NMR spectrum also showed assayed to locate active zones by placing them in signals corresponding to Ar-OH ( 8 168.10, sterile polypropylene food trays and covering 166.46), Ar-H ( 8 152.18, 112.04, 108.34, 103.78) them with molten agar previously seeded with and Ar-Me (8 25.27). Consideration of the the test organism. All bioassay plates were incu­ above data suggested a molecular formula of bated at 25° or 30°C for 2 - 3 days, and the di­ C10H10O5 and a structural formula correspond­ ameter of any zones of inhibition recorded. ing to methyl,3-aldo-2:4-dihydroxy-6-methyl ben­ The effect of methyl haematommate on zoate which has the trivial name of methyl hae­ spore germination and mycelial growth in liquid matommate (Asahina & Shibata 1954a). culture was assessed as described previously (Muir et al 1982). OH

RESULTS CHO

Preliminary bioassays of crude methanolic extracts (Fraction I) revealed a broad spectrum of activity against dermatophyte fungi, whilst di­ HX rect bioassay of TLC plates indicated two zones of antibiotic activity at Rf (0.82 and 0.65), with COOCH3 maximum activity in the former (= Compound SPECTRUM OF ANTIBIOTIC ACTIVITY The antibiotic activity of methyl hae­ In order to isolate Compound I a prelimi­ matommate was assayed against a range of bac­ nary partitioning was carried out by washing the teria and fungi. It appeared to show preferential methanolic extract (Fraction I) with hexane. The antibiotic activity against dermatophytes pooled hexane extracts were evaporated to (Table 1). Repetition of these bioassays by liq­ dryness in vacuo and then extracted with three uid culture assays and slide germination tests successive aliquots of acetone which removed the (Muir et al. 1982) confirmed these results, and active principle. Following repeated recrystalli­ showed that methyl haematommate was active sation from acetone-hexane (3:1), 50 g of dried against both spores and growing mycelium. lichen finally yielded 23 mg of pale-yellow, long, needle-shaped crystals, M.Pt 1470. MICROSCOPY Identification of the structure of Compound Examination of methyl haematommate- I was achieved by means of UV, 13C and aH- treated macroconidia of M. gypseum revealed NMR spectroscopy. The UV spectrum in marked disruption of the cytoplasmic organisa­ methanol showed peaks at 207, 241, 262 and 336 tion and thickening of the cell wall and mem­ nm. Addition of sodium acetate caused brane. Methyl haematommate-treated spores bathochromic shifts to 213, 272, and 378 nm, failed to germinate after 14 days. whilst in the presence of sodium methoxide Further examination of treated spores using BJ. Hickey et al.: Anti-fungal agent from S. ramulosum 51

Table 1. Antimicrobial activity of methyl haematommate (MH) as determined by AA disc bioassay. Numbers in brackets • number of strains tested.

DIA1VIETE R OF ZONE OF INHIBITION (mm) ORGANISM Discs apphed before Discs apphed after germination germination

MH (^g/disc) 340 34 3.4 0.34 340 34 3.4

Dermatophytes Epidermophyton floccosum (3) 11-12 5-8 2 - 5-6 3-4 1 Microsporum canis (2) 17-18 7-9 2 - 5-6 1 - Af. nanum (1) 15 10 3 - 5 2 - Af. gypseum (1) 19 14 4 1 9 5 1 Trichophyton rubrum (3) 14-17 7-8 - - 5 2 - T. mentagrophytes (2) 13-16 8-11 3-4 - 7-9 2-6 1 T. mentagrophytes var interdigitale (2) 14-15 8-11 - - 6-7 2 -

Non dermatophytes Aspergillus fumigatus ------A. niger ------Bacillus subtilis ------Candida albicans - - - - 3 3 1 Cladosporium resinac 4 3 ' - - - - - Escherichia coli ------Fusarium oxysporum 4 - - - - -' - Pseudomonas aeruginosa ------Staphylococcus aureus ------Verticillium dahliae 15 ll 3 - 3 - -

transmission electron microscopy showed con­ same Rf and colour reactions as ergosterol, siderable loss of internal organisation, and dis­ whereas the methyl haematommate-treated cells ruption of the cell membrane. Control (un­ showed loss of the ergosterol spot, and gained a

treated) spores had intact cytoplasm and mem­ new spot with the same Rf as squalene. branes, and many had produced germ tubes. DISCUSSION STEROL ANALYSIS A number of potent anti-fungal agents are produce an extraordinarily diverse known to affect ergosterol biosynthesis in fungi range of secondary metabolites whose exact role (Nozawa & Morita 1986). In preliminary tests E. in nature is still unclear. Many have antibiotic floccosum, T. mentagrophytes and M. canis were properties. Lichen extracts have been used as a grown on Sabouraud agar containing a sub-lethal component of the folk medicines of a number of concentration (17mg ml'1) of methyl haematom­ peoples (Asahina & Shibata 1954b, Richardson mate. Their sterols were then extracted and com­ 1988). Lichen acids have been shown to retard pared by TLC to those extracted from control the growth of Gram-positive (Vartia mycelium grown in the absence of methyl hae­ 1973) whilst more recently Ingolfsdottir et al. matommate. Chromatograms of the untreated (1985) screened Icelandic lichens for antibiotic dermatophytes exhibited spots possessing the activity and showed that the products of hydroly- 52 New Zealand Natural Sciences 17 (1990)

sis of certain depsides were active against bacte­ REFERENCES ria and fungi. They identified the active com­ pounds as methyl j3 -orsellinate and the methyl Asahina, Y. and Shibata, S. (1954a). Chemistry of and ethyl esters of orsellinic acid, and suggested Lichen Substances, p95. Japan Society for that these esters were hydrolysis products of the Promotion of Science. Tokyo. depsides atranorin and tenuiorin. Asahina, Y. & Shibata, S. (1954b). Chemistry of In this paper we have shown that methyl hae­ Lichen Substances. Japan Society for Promo­ matommate from S. ramulosum exhibits antibi­ tion of Science. Tokyo, Chapter 3, pp216- otic activity against dermatophyte fungi, proba­ 224. bly by affecting steroid biosynthesis and the or­ Calder, V.L., Cole, A.LJ. & Walker, J.R.L. ganisation of membrane structure. Haematom- (1986). Antibiotic compounds from N.Z. plants: III, a survey of N.Z. plants for antibi­ mic acid, or its esters, does not normally occur as otic substances. Journal Royal Society of such in lichens but it is found as a component of New Zealand 16:169-181. common lichen depsides such as atranorin and Culberson, CF. & Kristinsson, H. (1970). A norstictic acid and from which it may be ob­ standardised method for identification of li­ tained by alcoholysis (Asahina & Shibata 1954a). chen products. Journal of Chromatography Both of these depsides occur in S. ramulosum 46: 85-93 (Lamb 1951) and it is possible that our isolation Culberson, CF. (1969). Chemical and Botanical of methyl haematommate could be an artefact of Guide to Lichen Products p23. University of the isolation procedure (Hylands & Ingolfsdottir North Carolina Press, Chapel Hill, U.S A. 1985). However when small portions of intact Hylands, P.J. & Ingolfsdottir, K. (1985). The iso­ thallus were subjected to agar plate bioassay, lation of methyl j8 -orsellinate from Stere­ zones of inhibition were observed around them; ocaulon alpinum and comments on the isola­ thus it is probable that free methyl haematom­ tion of 4:6-dihydroxy-2-methoxy-7-methy- mate was present in S. ramulosum. Culberson lacetophenone from Stereocaulon species. (1969) suggested that these depsides are formed Phytochemistry 24:127-129. by the phycobiont from aromatic precursors syn­ Ingolfsdottir, K., Bloomfield, S.F. & Hylands, thesised by the mycobiont. By analogy with gly­ PJ. (1985). In vitro evaluation of the antimi­ coside formation in higher plants this synthesis crobial activity of lichen metabolites as po­ could serve as an internal protective mechanism tential preservatives. Antimicrobial Agents for the algal partner. Lichens also secrete en­ and Chemotherapy 28: 289-292. zymes that can hydrolyse depsides to release free Lamb, LM. (1951). On the morphology, phylo­ phenolic compounds which might act in chemical geny and of the lichen genus Stere­ defence mechanisms (Mosbach & Ehrensvaard ocaulon. Canadian Journal of Botany 29: 1966). 522-584. Lisboa, B.P. (1969). In Methods in Enzymology (ed. R.B. Clayton) Vol 15, pp3-157. Aca­ ACKNOWLEDGEMENTS demic Press, New York. Mosbach, K. and Ehrensvaard, U. (1966). Stud­ The authors wish to thank Dr J.W. Blunt and ies on lichen enzymes. I. Preparation and Dr M.H.G. Munro for assistance with the NMR properties of a depside hydrolysing esterase analyses, Mr Ron Bridger (Godfrey Pathology and of orsellinic acid decarboxylase. Bio­ Lab) and Messrs G. Paltridge and B. Harris chemical and Biophysical Research Commu­ (Christchurch Public Hospital) for clinical iso­ nications 22:145-150.. lates of microorganisms, the University Grants Muir, A.D., Cole, A.LJ. & Walker, J.R.L. Committee and the University of Canterbury for (1982). Antibiotic compounds from New the provision of research equipment and facili­ Zealand plants. I; Falcarindiol, an anti-der- ties. Mrs Margaret Stevens provided competent matophyte agent from Schefflera digitata. Journal of Medicinal Plant Research 44: 129- technical assistance. 133. BJ. Hickey et al.: Anti-fungal agent from S. ramulosum 53

Nozawa, Y. & Morita, T. (1986). In In vitro and Richardson, D.H.S. (1988). In CRC Handbook of In vivo Evaluation of Antifungal Agents, (ed. Lichenology. (ed. M. Galun), Vol III. ppl03- K. Iwata & H. Vanden Bossche), pplll-122, 105. CRC Press Inc. Boca Raton. Florida, Elsevier Science Publishers B.V. Amster­ U.SA. dam. Stahl, E. (Ed.) (1969). Thin Layer Chromatogra­ Parkes, L.W., Botlema, C.D.K., Rodriguez, RJ. phy, 2nd edition. George Allen & Unwin & Lewis, TA. (1985). In Methods in Enzy­ Ltd. London. mology (ed. J.H. Law & H.C. Rilling), Vol Vartia, K.O. (1973). In The Lichens, (ed. V. Ah- lll. pp333-345. Academic Press Ltd. New madjian and M.E. Hale), ppl45-150. Aca­ York. demic Press, New York.